We are creating a unified UKRI website that brings together the existing research council, Innovate UK and Research England websites.
If you would like to be involved in its development let us know.

Site search
Back to blog

Clinical trials: why multi-arms are better than two

by Guest Author on 25 Jul 2014

A pile of blue and white pillsIn their quest to speed up the discovery of new treatments for patient benefit, MRC researchers at the MRC Clinical Trials Unit propose an efficient solution: the multi-arm clinical trial. But how do these trials work and why should the traditional two-arm clinical trial be the exception and not the rule? We asked Annabelle South, Policy, Communications & Research Impact Coordinator at the MRC Clinical Trials Unit, to explain this change in approach, as published in The Lancet.

Randomised controlled trials are the best way of finding out if new treatments are better than the current standard treatment. But trials take a long time to set up and run, and can be expensive.

Here at the MRC Clinical Trials Unit at UCL we are keen to find faster and more efficient ways of working out which treatments work best, so patients can benefit sooner. That’s why we are urging researchers and drugs companies to move away from the traditional ‘two-arm’ approach to trials, and adopt more innovative, efficient approaches like the multi-arm trial.

The usual design for a randomised controlled trial is to randomly allocate volunteer patients to receive either the new treatment or the current standard one. The trial doctors and nurses closely monitor both groups (or arms). At the end of the trial, we analyse all the data to see if the new treatment works better than the old one.

This design does work – it tells us whether the new treatment is effective and safe. But about half of trials find the new treatment is no better than the existing one, leaving us back at the drawing board.

We believe that multi-arm trials can offer advantages over two-arm trials, and here’s how:

  1. Increase your chance of finding an effective treatment

If you increase the number of new treatments you are testing in a trial, you increase the chances that one of them will be better. This means that at the end of all your efforts you are much more likely to have discovered something that can improve the lives of people with the disease.

  1. Test more treatments at once

Multi-arm trials allow you to test many more treatments than would ever be done in two-arm trials – you don’t have to guess in advance which new drug is most likely to work, you can test several good candidates at once.

  1. Halve your costs

It’s much cheaper to assess several new treatments simultaneously in one multi-arm trial, rather than run several separate trials – we estimate that costs are halved.

  1. Speed up recruitment of participants

Patients and doctors support it. Some people have been concerned that multi-arm trials may be too complex for doctors and patients to understand. Our experience shows this is not the case – in fact, patients and doctors are very supportive of multi-arm trials, asking “why don’t you do all your trials this way?” Our multi-arm trials recruit participants faster than comparable two-arm trials.

Going one step further

Multi-arm, multi-stage (MAMS) trials take this one step further. They not only test several treatments at once, but can also stop recruiting patients to a particular arm part way through, if a treatment is not looking promising enough. They even allow us to add new treatments to the trial when they are ready for testing. This approach is even more efficient, meaning you don’t have to have gaps between stopping a trial and starting a new one, and may not need to set up a separate new trial to test a new treatment. This saves many years.

One example of this is the STAMPEDE trial, looking at how to treat prostate cancer. It has stopped and added arms as it’s gone along. It now has more than 5,500 participants, making it the largest prostate cancer treatment trial ever. It will assess eight new treatments in 15 years – something that would have taken at least 40 years in separate sequential trials.

The advantages of multi-arm trials over two-arm trials are such that two-arm trials should be the exception, not the rule. We urge doctors and patient groups to push for more multi-arm trials, and fewer two-arm trials, so we can identify effective new treatments faster and more efficiently.

Annabelle South

Further reading

More multiarm randomised trials of superiority are needed (sign in required)

Scientists call for more multi-arm clinical trials to speed up approval of new drugs

The road to randomisation: patulin and the common cold


Clinical research enables researchers and doctors to find new and better ways to understand, detect, control and treat human diseases. A clinical research study is a way to find answers to difficult scientific or health questions. As I am student of clinical research. I found your article very helpful for my project research. Thanks for putting out top notch content all the time. I would like to be here again to find another masterpiece article.

author avatar by Clinical research on 23-Mar-2019 03:53

the very first and important role of that Clinical researchers are responsible for ensuring these trials are conducted in an ethical manner whilst following good clinical practice and ensuring all team members are compliant with strict rules and regulations.

author avatar by Ridhima Patel on 18-Nov-2019 05:02

Randomized trials definitely help clinical trials as they improve the quality of health in patients over a period of time.

author avatar by Nisha on 04-Feb-2020 05:39

The abovementioned issue of relatedness between the individual questions should not be confused with statistical correlation between the corresponding statistical tests associated with each question ( 20 – 22 ). In a multi-arm study, individual comparisons are positively correlated due to the use of the same control arm (this is not the case for separate trials). Because of this, a multi-arm trial has a lower overall probability of any false-positive result but a higher probability of making more than one false-positive conclusion (relative to separate trials; ref. 9 ). However, these probability differences are small ( 9 ), especially when the number of experimental arms is in a practical range (two to four arms). Therefore, the fundamental issue for the purpose of multiplicity adjustment is the relatedness of clinical questions with the statistical correlation having minimal relevance. RCTs incorporate formal interim-monitoring guidelines to allow stopping early for strong evidence of benefit of the experimental agent (efficacy) or when it seems that the experimental agent will not be better than the control treatment (futility). The presence of

author avatar by Eaton Rosalyn on 21-Oct-2020 10:08

Leave a reply

You may use basic HTML in your comments. Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.


From category

Share this: